Dominance of virus over host factors in cross-species activation of human cytomegalovirus early gene expression

Abstract

Human cytomegalovirus (HCMV) exhibits a highly restricted host range. In this study, we sought to examine the relative significance of host and viral factors in activating early gene expression of the HCMV UL54 (DNA polymerase) promoter in murine cells. Appropriate activation of the UL54 promoter at early times is essential for viral DNA replication. To study how the HCMV UL54 promoter is activated in murine cells, a transgenesis system based on yeast artificial chromosomes (YACs) was established for HCMV. A 178-kb YAC, containing a subgenomic fragment of HCMV encompassing the majority of the unique long (UL) region, was constructed by homologous recombination in yeast. This HCMV YAC backbone is defective for viral growth and lacks the major immediate-early (IE) gene region, thus permitting the analysis of essential cis-acting sequences when complemented in trans. To quantitatively measure the level of gene expression, we generated HCMV YACs containing a luciferase reporter gene inserted downstream of either the UL54 promoter or, as a control for late gene expression, the UL86 promoter, which directs expression of the major capsid protein. To determine the early gene activation pathway, point mutations were introduced into the inverted repeat 1 (IR1) element of the UL54 promoter of the HCMV YAC. In the transgenesis experiments, HCMV YACs and derivatives generated in yeast were introduced into NIH 3T3 murine cells by polyethylene glycol-mediated fusion. We found that infection of YAC, but not plasmid, transgenic lines with HCMV was sufficient to fully recapitulate the UL54 expression program at early times of infection, indicating the importance of remote regulatory elements in influencing regulation of the UL54 promoter. Moreover, YACs containing a mutant IR1 in the UL54 promoter led to reduced ( approximately 30-fold) reporter gene expression levels, indicating that HCMV major IE gene activation of the UL54 promoter is fully permissive in murine cells. In comparison with HCMV, infection of YAC transgenic NIH 3T3 lines with murine cytomegalovirus (MCMV) resulted in lower (more than one order of magnitude) efficiency in activating UL54 early gene expression. MCMV is therefore not able to fully activate HCMV early gene expression, indicating the significance of virus over host determinants in the cross-species activation of key early gene promoters. Finally, these studies show that YAC transgenesis can be a useful tool in functional analysis of viral proteins and control of gene expression for large viral genomes.

FIG. 1

(A) Transient transfection assays using 1 μg of DNA of the plasmids described, followed by infection with HCMV at an MOI of 10. NIH 3T3 cells were lysed at different times postinfection or at the moment of infection (mock), and luciferase activity of the extracts was measured. Relative luciferase activities from triplicate measurements are shown. The RLU values at each time postinfection are normalized to the activity in the corresponding mock-infected cells. (B) Stably transfected clones were infected with HCMV at an MOI of 10, and samples were taken at different times postinfection. The average RLU values from at least two experiments performed in triplicate are shown. Clones 2A-1, 2A-3, and 2A-5 are three representative clones stably transfected with pLHN.UL54wt; 6A-1, 6A-2, and 6A-4 are representative clones stably transfected with pLHN.UL86.

FIG. 2

Structure of the HCMV genome, showing the UL and US (unique short) regions and the inverted and direct repetitive sequences. The STS diagram shows relative positions of the pairs of oligonucleotides used to check the integrity of HCMV. Five YACs are shown, with their corresponding mutations outlined. The selectable markers (U, URA3; T, TRP1; L, LEU2; P, puromycin resistance; Neo, neomycin resistance) are shown, as well as the positions of the HCMV sequences where the mutations were performed.

FIG. 3

(A) PFGE to resolve different DNA samples: size markers (lane 1), YPH857 (lanes 2 and 4), Y24 (lanes 3, 5, and 6), and Y24P (lane 7). The gels were either stained with ethidium bromide (lanes 1 to 3) or hybridized to a UL54-specific probe (pol; lanes 4 and 5) or a puromycin-specific probe (puro; lanes 6 and 7). Sizes of the marker DNA fragments are shown at the left; arrows indicate bands corresponding to either Y24 or Y24P. (B) Agarose gel in which an aliquot of the STS PCR mixture was loaded. The lanes are marked V for AD169 DNA or Y for YAC DNA. The set numbers for the corresponding STS, as well as locations of the size markers, are indicated. (C) PCR to test the UL54-Luc and UL86-Luc mutations and their integration in the correct position in Y24P. The specific primers used were YAC54.2 and GLprimer2 (lanes 1 to 3), and YAC86.2 and GLprimer2 (lanes 4 and 5). The DNA samples are Y24P (lanes 1 and 4), Y24P/UL54wt-Luc (lane 2), Y24P/UL54IRM-Luc (lane 3), and Y24P/UL86-Luc (lane 5).

FIG. 4

(A) Y24P/UL54wt-Luc transgenic lines were infected with HCMV at an MOI of 10. Samples were harvested at the indicated times postinfection. Averages of at least two experiments done in triplicate are shown. As a negative control, Y24P/UL86-Luc clones 5C and 7E were used. (B) Fold activation of luciferase activity in transgenic plasmid pLHN.UL54wt lines 2A-1, 2A-3, and 2A-5 and in Y24P/UL54wt-Luc lines 31C, 42.8, and 44.11.

FIG. 5

The stable NIH 3T3 line containing Y24P/UL54wt-Luc clone 44.11 was infected with HCMV at MOIs from 0.005 to 10. Cells were harvested at 48 hpi, and luciferase activity was measured. Each point represents the average of two independent infections.

FIG. 6

STS analysis of YAC transgenic NIH 3T3 clones. PCRs were performed using specific primer sets designed along the HCMV genome (see Materials and Methods). DNA from NIH 3T3 cells (lanes 1, 4, 7, 10, and 13), yeast Y24P positive control (lanes 2, 5, 8, 11, and 14), or NIH 3T3 cells expressing Y24P/UL54wt-luc clone 44.11 (lanes 3, 6, 9, 12, and 15) was used, along with different sets of primers as indicated at the top.

FIG. 7

Stable Y24P/UL54IRM-Luc-containing NIH 3T3 clones were infected with HCMV at an MOI of 10. Samples were harvested at the indicated times, and luciferase activity determined as indicated in Materials and Methods. Each point represents the average of at least two independent experiments done in triplicate. Fold activation was calculated by dividing the absolute value by the mock-infected value. Results for three representative clones are shown.

FIG. 8

Detection of HCMV IE1 and UL54 transcripts in 3T3 cells infected with HCMV. NIH 3T3 or HFF cells were mock infected (M) or infected with HCMV at an MOI of 1. Total RNA was harvested at different times (hours postinfection) indicated, treated with DNase, and reverse transcribed by using oligo(dT). PCRs were performed using primer sets specific for HCMV IE1 UL54, human TF (for HFF samples), and murine HPRT (for NIH 3T3 samples) as described in Materials and Methods. Amplified products were separated on 1% agarose gels and visualized by ethidium bromide staining. Amplified fragments obtained in the different reactions are shown. Sizes were as expected for each primer set (see Materials and Methods for details). Specific PCR-amplified products were not detected in control reactions in which reverse transcriptase was not added during the RNA reverse transcription reaction (data not shown).

FIG. 9

Y24P/UL54wt-Luc-containing NIH 3T3 clones were infected with MCMV at an MOI of 10. Samples were harvested at the indicated times, and luciferase activity measured. Each point represents the average of at least two independent experiments done in triplicate. Fold activation was calculated by dividing the absolute value by the mock-infected value. Results for two representative clones are shown.